In Project 3, we propose to use a rat model to characterize the neural substrates of cocaine-induced alterations in fMRI signals. An animal model is required because the invasive methods used cannot be applied to human subjects. The proposed experiments will characterize the underlying biological processes that contribute to the fMRI signal, including an analysis of contributions due to the direct effects of cocaine on the vascular versus its direct effects on target neurons. We then plan to investigate the postsynaptic correlates of the fMRI signal by comparing patterns of regional brain activation obtained by fMRI with in situ hybridization for the immediate early genes (IEGs) c-fos and zif268. We propose this approach because mapping of IEG induction is the best current functional neuroanatomic tool to investigate circuits, cells, and neurotransmitter receptors involved in cocaine action. In addition we propose to identify basic mechanisms by which cocaine induces both fMRI and IEG 'signals' by a series of studies in which the monoamine systems acted upon by cocaine are lesioned. The experiments proposed in Project 3 directly impact our interpretation of the human data obtained in Project 1. Project 3 is designed to rigorously test hypotheses concerning the biological significance of fMRI signals produced by cocaine administration. At the same time it should produce significant data on the neural substrates of cocaine dependence.